Traditional and recent approaches in background modeling for foreground detection: An overview

Recent research on problem formulations based on decomposition into low-rank plus sparse matrices shows a suitable framework to separate moving objects from the background. The most representative problem formulation is the Robust Principal Component Analysis (RPCA) solved via Principal Component Pursuit (PCP) which decomposes a data matrix in a low-rank matrix and a sparse matrix. However, similar robust implicit or explicit decompositions can be made in the following problem formulations: Robust Non-negative Matrix Factorization (RNMF), Robust Matrix Completion (RMC), Robust Subspace Recovery (RSR), Robust Subspace Tracking (RST) and Robust Low-Rank Minimization (RLRM). The main goal of these similar problem formulations is to obtain explicitly or implicitly a decomposition into low-rank matrix plus additive matrices. In this context, this work aims to initiate a rigorous and comprehensive review of the similar problem formulations in robust subspace learning and tracking based on decomposition into low-rank plus additive matrices for testing and ranking existing algorithms for background/foreground separation. For this, we first provide a preliminary review of the recent developments in the different problem formulations which allows us to define a unified view that we called Decomposition into Low-rank plus Additive Matrices (DLAM). Then, we examine carefully each method in each robust subspace learning/tracking frameworks with their decomposition, their loss functions, their optimization problem and their solvers. Furthermore, we investigate if incremental algorithms and real-time implementations can be achieved for background/foreground separation. Finally, experimental results on a large-scale dataset called Background Models Challenge (BMC 2012) show the comparative performance of 32 different robust subspace learning/tracking methods.

Decomposition into Low-rank plus Additive Matrices for Background/Foreground Separation: A Review for a Comparative Evaluation with a Large-Scale Dataset

https://hal.archives-ouvertes.fr/hal-01522823/document

Decomposition into low-rank plus additive matrices for background/foreground separation

Ship detection plays an important role in automatic remote sensing image interpretation. The scale difference, large aspect ratio of ship, complex remote sensing image background and ship dense parking scene make the detection task difficult. To handle the challenging problems above, we propose a ship rotation detection model based on a Feature Fusion Pyramid Network and deep reinforcement learning (FFPN-RL) in this paper. The detection network can efficiently generate the inclined rectangular box for ship. First, we propose the Feature Fusion Pyramid Network (FFPN) that strengthens the reuse of different scales features, and FFPN can extract the low level location and high level semantic information that has an important impact on multi-scale ship detection and precise location of dense parking ships. Second, in order to get accurate ship angle information, we apply deep reinforcement learning to the inclined ship detection task for the first time. In addition, we put forward prior policy guidance and a long-term training method to train an angle prediction agent constructed through a dueling structure Q network, which is able to iteratively and accurately obtain the ship angle. In addition, we design soft rotation non-maximum suppression to reduce the missed ship detection while suppressing the redundant detection boxes. We carry out detailed experiments on the remote sensing ship image dataset, and the experiments validate that our FFPN-RL ship detection model has efficient detection performance.

A Ship Rotation Detection Model in Remote Sensing Images Based on Feature Fusion Pyramid Network and Deep Reinforcement Learning

In this paper, we propose a sparse coding approach to background modeling. The obtained model is based on dictionaries which we learn and keep up to date as new data are provided by a video camera. We observe that, without dynamic events, video frames may be seen as noisy data belonging to the background. Over time, such background is subject to local and global changes due to variable illumination conditions, camera jitter, stable scene changes, and intermittent motion of background objects. To capture the locality of some changes, we propose a space-variant analysis where we learn a dictionary of atoms for each image patch, the size of which depends on the background variability. At run time, each patch is represented by a linear combination of the atoms learnt online. A change is detected when the atoms are not sufficient to provide an appropriate representation, and stable changes over time trigger an update of the current dictionary. Even if the overall procedure is carried out at a coarse level, a pixel-wise segmentation can be obtained by comparing the atoms with the patch corresponding to the dynamic event. Experiments on benchmarks indicate that the proposed method achieves very good performances on a variety of scenarios. An assessment on long video streams confirms our method incorporates periodical changes, as the ones caused by variations in natural illumination. The model, fully data driven, is suitable as a main component of a change detection system.

Online Space-Variant Background Modeling With Sparse Coding

A Vehicle Detection Algorithm Based on Compressive Sensing and Background Subtraction

Moving-object Detection Based on Sparse Representation and Dictionary Learning

Provided is a real-time transport vehicle detecting and tracking method. According to the real-time transport vehicle detecting and tracking method, in moving vehicle real-time detection, vehicle traveling areas are obtained by utilizing road line detection, main color information of moving areas and main color information of non-moving areas in the vehicle traveling areas are obtained by carrying out frame subtraction, the moving vehicle areas are filled with the main color information to obtain an approximate background image, and finally moving vehicles are obtained by utilizing background subtraction. In moving vehicle tracking, all feature angle points of the moving vehicles are obtained by utilizing Harris detection, feature angle point sets of all separated moving areas are obtained through clustering analysis, each feature angle point set generates a feature circle containing all feature points, problems of vehicle blocking and the like are analyzed by using radiuses of the feature circles, and finally feature matching tracking is carried out by using circle centers of the feature circles. According to the real-time transport vehicle detecting and tracking method, moving vehicle detection, instantaneity of tracking and tracking precision are all improved greatly, and the real-time transport vehicle detecting and tracking method has wide application significance.

Real-time transport vehicle detecting and tracking method

This paper proposed methods of vehicle detection and tracking algorithm in real-time traffic. In the detection of realtime moving vehicle, vehicle areas would be determined through road line detection. Then, the main color information of moving and non-moving area would be obtained through frame difference. Filling the main color information in vehicle moving area would lead to a similar background image. At last, moving vehicles would be determined through adaptive Background Subtraction difference. In the tracking of moving vehicles, firstly, all characteristic corners can be got by using Harris detection. Then, all characteristic corner set in the separate moving area would be collected through cluster analysis. For each characteristic individual corner set can generate a circle embracing all characteristics, some problems like vehicle barrier could be analyzed by using the radius of characteristic circle. At last, conduct feature matching tracking by using the center of feature circle. Experimental result shows that the improving algorithm can extract all moving objects, which was endowed with strong background adaptability and better real-time performance. Keywords-target detection; frame difference method; background subtraction difference method; harris corner detection; clustering analysis

/pdf/research-on-vehicle-detection-and-tracking-algorithm-based-53z9vv0le3.pdf

Research on Vehicle Detection and Tracking Algorithm Based on the Methods of Frame Difference and Adaptive Background Subtraction Difference

Most state-of-art deep learning based no reference image quality assessment (NR IQA) methods usually have a complex network structure which are memory-consuming, hard to train and so not applicable in practical scenarios. Aim at these problems, a RepVGG based NR-IQA algorithm with transfer learning is proposed. The method uses ImageNet dataset to pre-train RepVGG network to get network parameters, and then uses the trained network to extract image features of image quality assessment data set. Finally, a simple fully connected network is trained to get the quality score of the image based on these features. The experimental results show that on the KADID-10K, LIVE、TID2013 and CSID datasets, the overall objective assessment obtained by the method is better than the state-of-art deep learning-based methods with good consistency with the subjective assessment. 

Xiaosheng Huang

Papers

A Vehicle Detection Algorithm Based on Compressive Sensing and Background Subtraction

Moving-object Detection Based on Sparse Representation and Dictionary Learning

Real-time transport vehicle detecting and tracking method

Research on Vehicle Detection and Tracking Algorithm Based on the Methods of Frame Difference and Adaptive Background Subtraction Difference

A no reference image quality assessment method based on RepVGG